Download Effective Elimination of Cancer Stem Cells By a Novel Drug

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
CANCER STEM CELLS
Effective Elimination of Cancer Stem Cells By a Novel Drug
Combination Strategy
SHUQIANG YUAN,a,b FENG WANG,a,b GANG CHEN,b HUI ZHANG,b LI FENG,b LEI WANG,c HOWARD COLMAN,d
MICHAEL J. KEATING,e XIAONAN LI,f RUI-HUA XU,a JIANPING WANG,c PENG HUANGa,b
a
State Key Laboratory of Oncology in South China, Sun Yat-sen University Cancer Center, Guangzhou,
Guangdong, China; bDepartment of Molecular Pathology; and eDepartment of Leukemia, The University of Texas
MD Anderson Cancer Center, Houston, Texas, USA; cDepartment of Colorectal Surgery, Sixth Hospital of Sun
Yat-sen University, Guangzhou, Guangdong, China; dDepartment of Neuro-Oncology, University of Utah, Salt
Lake City, Utah, USA; fDepartment of Molecular and Cellular Biology, Texas Children’s Hospital, Baylor
College of Medicine, Houston, Texas, USA
Key Words. Glioma • Hypoxia • Chemotherapy • Drug target • Toxicity
ABSTRACT
Development of effective therapeutic strategies to eliminate
cancer stem cells, which play a major role in drug resistance
and disease recurrence, is critical to improve cancer treatment outcomes. Our study showed that glioblastoma stem
cells (GSCs) exhibited low mitochondrial respiration and
high glycolytic activity. These GSCs were highly resistant to
standard drugs such as carmustine and temozolomide
(TMZ), but showed high sensitivity to a glycolytic inhibitor
3-bromo-2-oxopropionate-1-propyl ester (3-BrOP), especially under hypoxic conditions. We further showed that
combination of 3-BrOP with carmustine but not with TMZ
achieved a striking synergistic effect and effectively killed
GSCs through a rapid depletion of cellular ATP and inhibition of carmustine-induced DNA repair. This drug combination significantly impaired the sphere-forming ability of
GSCs in vitro and tumor formation in vivo, leading to
increase in the overall survival of mice bearing orthotopic
inoculation of GSCs. Further mechanistic study showed that
3-BrOP and carmustine inhibited glyceraldehyde-3-phosphate dehydrogenase and caused a severe energy crisis in
GSCs. Our study suggests that GSCs are highly glycolytic
and that certain drug combination strategies can be used to
effectively overcome their drug resistance based on their
metabolic properties. STEM CELLS 2013;31:23–34
Disclosure of potential conflicts of interest is found at the end of this article.
INTRODUCTION
Cancer stem cells (CSCs) are a subgroup of cancer cells that
have the ability to self-renew, to differentiate into multiple lineage cells, and to initiate tumors in vivo [1]. They have been
found in hematopoietic malignancies [2] and different types of
solid tumors including brain [3], breast [4], colon [5], and pancreatic [6] cancers. A growing body of studies indicates that
CSCs are intrinsically more resistant to chemotherapeutic
agents and radiation than the bulk of tumor cells, and thus play
an important role in persistence of cancer residual disease and
recurrence [1]. This drug resistance in CSCs has been attributed to highly expressed drug efflux pumps (such as multidrug
resistance proteins), enhanced DNA repair proteins, expression
of antiapoptotic proteins, and a slow rate of cell proliferation
[1]. Thus, it is important to develop effective therapeutic strategies to eliminate CSCs and overcome cancer resistance to
chemotherapy and radiotherapy. However, currently very lim-
ited therapeutic strategies are effective in eliminating CSCs,
which remains a major challenge in cancer treatment.
Glioblastoma multiforme (GBM), a WHO grade IV astrocytoma, is the most common and aggressive primary brain tumor in adults. Although maximal surgical resection, radiotherapy, and chemotherapy are performed in GBM patients, the
treatment outcomes are still dismal, with a median survival of
only 12–15 months and the 5-year survival rate of less than
10% [7, 8]. Previous studies demonstrated that glioblastoma
stem cells (GSCs) are resistant to conventional chemotherapy
drugs carmustine (bis-chloroethylnitrosourea, BCNU) and
temozolomide (TMZ) as well as radiation [9, 10]. Since the
GSCs are probably responsible for the recurrence of GBM
[11–14], how to target the GSCs became a crucial question.
The GSCs have been found in the hypoxic niches, which further promote drug resistance [15–17]. Under hypoxic conditions, cancer cells are more dependent on the glycolytic pathway to generate ATP and metabolic intermediates for survival
and proliferation. Based on these observations, we postulated
Author contributions: S.Y. and F.W.: conception and design, collection/assembly of data, data analysis/interpretation, manuscript writing,
and final approval; G.C., H.Z., L.W., H.C., M.K., R.X., and J.W.: provision of study materials and final approval; L.F.: collection/
assembly of data; X.L.: design of animal study and data analysis/interpretation; P.H.: conception and design, provision of study
materials, data analysis/interpretation, and final approval; S.Y. and F.W. contributed equally to this article.
Correspondence: Peng Huang, M.D., Ph.D., Department of Molecular Pathology, Unit 951, The University of Texas MD Anderson
Cancer Center, 1515 Holcombe Blvd., Houston, Texas 77030, USA. Telephone: 713-834-6044; Fax: 713-834-6084; e-mail: phuang@
mdanderson.org Received May 1, 2012; accepted for publication September 29, 2012; first published online in STEM CELLS EXPRESS
C AlphaMed Press 1066-5099/2012/$30.00/0 doi: 10.1002/stem.1273
November 6, 2012. V
STEM CELLS 2013;31:23–34 www.StemCells.com
Elimination of Glioblastoma Stem Cells
24
that GSCs might be more reliant on glycolysis to maintain their
energy homeostasis and stemness than non-stem tumor cells. As
such, targeting glycolytic pathway might be a preferential and
effective strategy to kill GSCs.
Development of novel therapeutic agents that target cancer
cell metabolism has become an important field of research.
Compounds known to inhibit the glycolytic pathway include 2deoxyglucose and 3-bromopyruvate (3-BrPA) [18–20]. In particular, 3-BrPA is an alkylating agent that has been shown to
inhibit hexokinase and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), two key enzymes in the glycolytic pathway
[18, 21]. A derivative of 3-BrPA, 3-bromo-2-oxopropionate-1propyl ester (3-BrOP), is chemically more stable than 3-BrPA
and has been shown to be highly potent in causing ATP depletion in cancer cells [22]. In this study, we found that GSCs
exhibited low mitochondrial respiration and high glycolytic activity, and further tested the possibility that 3-BrOP might be
able to effectively inhibit glycolysis in GSCs and cause severe
ATP depletion that might render GSCs incapable of repairing
DNA damage induced by chemotherapeutic agents. Using two
GSC cell lines, GSC11 and GSC23, which were established
from human primary glioblastoma tissues with high expression
of a stem cell marker CD133 [23], we showed that GSCs were
highly sensitive to 3-BrOP, especially under hypoxic conditions, and that combination of this compound with BCNU had
striking synergistic effect in eliminating the GSCs.
MATERIALS
AND METHODS
Chemicals and Reagents
BCNU, TMZ, and 3-BrPA were purchased from Sigma (St
Louis, MO, http://www.sigmaaldrich.com). 3-BrOP was synthesized by esterification of 3-bromo-2-oxopropionate (Sigma)
with 1-propanol (Sigma) as described previously [22].
Cells and Cell Cultures
GSC11 and GSC23 originally derived from human primary glioblastoma tissues were maintained in Dulbecco’s modified Eagle’s
medium (DMEM)/F-12 (Mediatech, Manassas, VA, http://
www.cellgro.com) supplemented with B-27 (Invitrogen, Carlsbad, CA, http://www.invitrogen.com), 2 mM glutamine (Mediatech), 20 ng/ml recombinant human epidermal growth factor
(R&D Systems, Minneapolis, MN, http://www.rndsystems.com),
and 20 ng/ml basic fibroblast growth factor (R&D Systems) [24].
To induce CSC differentiation, GSCs were cultured in DMEM/
F-12 medium containing 10% fetal bovine serum (FBS) for various periods of time as indicated in each experiment. The glioma
cell line U87 and nonmalignant human astrocytes (NHAs) were
maintained in DMEM (Mediatech) supplemented with 10% FBS.
Cells were seeded in culture flasks or plates overnight before
each treatment. To test the cytotoxic effect of drugs under
hypoxic conditions, cells were first preincubated in a chamber
with 2% oxygen (O2) and 5% carbon dioxide (CO2) for 18 hours
and then treated with the indicated compounds under the same
hypoxic conditions (2% O2) for the indicated time.
Cell Viability Assay
Cell-growth inhibition was assayed using a colorimetric assay
with 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)2-(4-sulfophenyl)-2H-tetrazolium (MTS) (Promega, Madison,
WI, http://www.promega.com). Briefly, GSCs were seeded in
96-well plates and then treated with indicated compounds at
various concentrations. After 72 hours incubation, 40 ll MTS
solution was added to each well and incubated for another 4
hours. The absorbance in each well at 490 nm was measured
using a Multiskan plate reader (Thermo Scientific, Waltham,
MA, http://www.thermoscientific.com). Cell apoptosis and necrosis were measured using flow cytometric analysis of cells
double-stained with FITC-annexin-V and propidium iodide
(PI). Briefly, samples were collected, dissociated, washed with
cold phosphate buffered saline (PBS), and suspended in the
annexin-V binding buffer. The cells were stained with annexinV for 15 minutes at room temperature, washed, and stained
with PI. The samples were analyzed using a FACSCalibur flow
cytometer equipped with the CellQuest Pro software program
(Becton-Dickinson, San Jose, CA, http://www.bd.com). Cell
growth was also confirmed by cell counting using a Coulter
Counter (Beckman, Brea, CA, https://www.beckmancoulter.
com). All metabolic parameters (O2 consumption, glucose
uptake, lactate production, ATP depletion, GAPDH, and hexokinase activity) were assessed prior to cell death and the results
were normalized by cell number.
Tumor Sphere-Forming Assay
GSCs were treated with indicated compounds for 3 hours and
then seeded in fresh medium in 96-well plates in a range of
10–1,000 cells per well; neurospheres were counted under a
microscope (Nikon, Tokyo, Japan, http://www.nikon.com) after 3 weeks of incubations.
Tumorigenicity in Orthotopic Mouse Model
All animal experiments were conducted at Baylor College of
Medicine according to an Institutional Animal Care and Use
Committee-approved protocol. The Rag2/severe complex
immune deficiency (SCID) mice, ages 8–12 weeks, were bred
and maintained in a specific pathogen-free animal facility.
GSC11 cells were cultured in GSC medium and made into
single cells by Accutase. The same number of GSC11 cells
were exposed to PBS, 20 lM BCNU, 20 lM 3-BrOP, or 20
lM BCNU plus 20 lM 3-BrOP in GSC medium at 37 C for
6 hours, then cells were collected, washed, and suspended in
serum-free culture medium. After mice were anesthetized
with sodium pentobarbital (50 mg/kg), GSC11 cells were inoculated into Rag2/SCID mice cerebral hemisphere (1 mm to
the right of the midline, 1.5 mm anterior to the lambdoid
suture, and 3 mm deep) at 2 105 cells per mouse via a 10ml 26-gauge Hamilton Gastight 1701 syringe needle. The animals were monitored daily without further drug treatment
until they developed signs of neurological deficit or became
moribund, at which time they were euthanized [25].
Glucose Uptake Assay
To compare glucose uptake by GSCs and the serum-induced differentiated GSCs (10% FBS, 20 days), the old culture medium
was removed from the respective cell culture, and the cells were
incubated in glucose-free, serum-free medium for 2 hours at
37 C and then incubated with [3H]2-deoxyglucose (0.4 lCi/ml)
for 60 minutes. After washing with cold PBS, each sample was
suspended in 0.5 ml of water and transferred to a vial with 3 ml
of scintillation fluid. The radioactivity was detected using a liquid scintillation machine (Beckman).
Oxygen Consumption Assay
To compare oxygen consumption by GSCs and the seruminduced differentiated GSCs (10% FBS, 20 days), the old cell
culture medium was removed from the respective samples,
and the cells were harvested and resuspended in warm serumfree medium pre-equilibrated with 21% oxygen. The cells (7
106 cells per milliliter) were placed in a sealed respiration
chamber equipped with a thermostat control and microstirring
device (Oxytherm; Hansatech Instruments, Norfolk, U.K.,
http://www.hansatech-instruments.com). Oxygen consumption
Yuan, Wang, Chen et al.
by the cells in the chamber was measured and constantly
monitored at 37 C using a Clark-type oxygen electrode and
was expressed as nanomols of O2 consumed as a function of
time (nmol/ml per minute).
Lactate Production Assay
To measure the lactate production by GSCs and the seruminduced differentiated GSCs (10% FBS, 20 days), cells were
plated in six-well plates and cultured in their respective medium for the indicated time. Aliquots of the culture medium
were analyzed using the Accutrend lactate analyzer (Roche, Indianapolis, IN, http://www.rocheusa.com/portal/usa/indianapolis).
To measure the effect of drug treatment on lactate production in
short-term culture, cells were plated in 96-well plates and treated
with the indicated compounds for 3 hours. A colorimetric lactate
assay kit was used according to the manufacturer’s protocol
(Eton Bioscience, San Diego, CA, http://www.etonbio.com). Absorbance in each well at 490 nm was measured using a Multiskan plate reader.
ATP Depletion Assay
GSCs were plated in 96-well plates and treated with the indicated compounds for 3 or 6 hours, 100 ll of mixed CellTiterGlo luminescent reagent (Promega) was then added to each
well and incubated for 10 minutes on an orbital shaker. Luminescence was measured using a Fluoroskan luminescence
scanner (Thermo Scientific).
Single-Cell Gel Electrophoresis Assay (Comet Assay)
After GSCs were treated with the indicated compounds for 3
hours, cell suspensions mixed with 0.5% low-melting-point
agar was placed onto a 24 50-mm slide precoated with 1%
regular agar. After the agar was solidified, slides were soaked
in a prechilled fresh lysing solution (2.5 M NaCl, 100 mM ethylenediaminetetraacetic acid, 10 mM Tris-HCl, 1% Triton X100, pH 10) for 1 hour at 4 C. After rinsing with 0.4 M Tris
buffer (pH 7.5), slides were placed in a reservoir filled with
fresh electrophoresis buffer (300 mM NaOH, 1 mM ethylenediaminetetraacetic acid, pH >13) for 15 minutes and then subjected to electrophoresis for another 15 minutes (25 V, 300
mA). Slides were then stained with CYBR Green I (Trevigen,
Gaithersburg, MD, http://www.trevigen.com) and photographed under a fluorescent microscope (Nikon). The percentage of tail DNA, which indicates damaged DNA, was analyzed
using the Casp software program (version 1.2.2) provided by
the CASPLab Comet Assay Project.
Western Blotting
Cellular proteins were separated using electrophoresis on 10%
sodium dodecyl sulfate polyacrylamide gel electrophoresis,
transferred to nitrocellulose membranes, and then blotted with
specific primary antibodies against H2AX, cH2AX (Upstate,
Billerica, MA, http://www.millipore.com), and b-actin. The
bound primary antibodies were detected using appropriate horseradish peroxidase-conjugated secondary antibodies, and the signals were detected using a SuperSignal enhanced chemiluminescence kit (Pierce, Rockford, IL, http://www.pierce.com).
GAPDH Enzymatic Activity Assay
Purified GAPDH enzyme from rabbit muscle (Sigma) or protein extracts prepared from GSC11 cells were used in the
GAPDH assays in vitro, using a GAPDH Assay Kit (ScienCell, Carlsbad, CA, http://www.sciencellonline.com) according to the manufacturer’s protocol. Purified GAPDH was
incubated in vitro with the indicated compounds (BCNU, 3BrOP) for 30 minutes and then added to a mixture of 6.67
mM 3-phosphoglyceric acid, 3.33 mM L-cysteine, 117 lM bwww.StemCells.com
25
NADH, 1.13 mM ATP, 3.33 U/ml 3-phosphoglycerate kinase,
and 100 mM triethanolamine buffer (pH 7.6). The change in
NADH fluorescence was then monitored using a Fluoroskan
spectrometer every minute for up to 20 minutes. For analysis
of GAPDH activity in cell extracts, GSC11 cells were first
incubated with or without 3-BrOP, BCNU, or their combination for 30 minutes, and protein lysates were prepared and immediately used for GAPDH activity assay without addition of
3-BrOP or BCNU in vitro.
Hexokinase Enzymatic Activity Assay
Lysates of GSCs pretreated with the indicated compound or
purified hexokinase II enzyme from Saccharomyces cerevisiae
(Sigma) were prepared as described in GAPDH enzymatic activity assay and added to a mixture of 6.5 mM MgCl2, 220
mM glucose, 2.7 mM ATP, 0.83 mM b-NAD, 0.24 U/ml glucose-6-phosphate dehydrogenase (Sigma), and 100 mM triethanolamine buffer (pH 7.6). Increases in NADH fluorescence
were measured using a Fluoroskan fluorescence scanner every
1 minute for up to 20 minutes.
Statistical Analysis
Data were analyzed using GraphPad Prism 5 (GraphPad
Software, La Jolla, CA, http://www.graphpad.com). Data
graphed with error bars represent mean and SEM from experiments done in triplicate unless otherwise noted. A two-sided
Student’s t test was used to determine the significance of difference between samples. Survival curves were compared
between different groups by log-rank test.
RESULTS
GSCs Exhibit Low Mitochondrial Respiration and
Are Resistant to TMZ and BCNU But Sensitive to
Glycolytic Inhibition
Two human GSC lines, GSC11 and GSC23, were used in this
study. Under stem cell culture conditions in serum-free medium, GSC11 cells exhibited typical neurosphere morphology
with high expression of neural stem cell marker nestin (Fig.
1A) and showed relatively low mitochondrial respiration as
evidenced by a low oxygen consumption rate (Fig. 1B).
When GSCs were induced to undergo differentiation by exposure to serum [26, 27], the cells gradually lost their neurosphere morphology in medium containing 10% FBS, became
attached to the culture dish as a monolayer, and exhibited a
substantial decrease in nestin expression (Fig. 1A). Associated
with this serum-induced differentiation phenotype, there were
also a decrease in other glioma stem markers such as CD133,
SOX2, and Notch1 (data not shown). Interestingly, mitochondrial respiration was substantially activated leading to approximately 75% increase in oxygen consumption (Fig. 1B, 1C),
accompanied by significant decreases in lactate production
(Fig. 1D) and a significant decrease in glycolytic index (Fig.
1E). A similar change in energy metabolism (decrease in glycolytic index) was also observed in GSC23 cells induced by
serum (Fig. 1E). Interestingly, the glycolytic index of the
‘‘regular’’ GBM cells (U87) was also lower than that of GSCs
in stem cell medium (Fig. 1E), suggesting that although regular cancer cells (non-stem) are generally considered active in
glycolysis, GSCs exhibit even higher glycolytic activity with
lower mitochondrial respiration, and serum induction activated mitochondrial function and reduced glycolysis.
When GSC11 cells were incubated with TMZ and BCNU
(two drugs commonly used in the treatment of glioblastoma)
26
Elimination of Glioblastoma Stem Cells
Figure 1. Low mitochondrial respiration and high glycolysis in GSCs and their resistance to standard chemotherapeutic agents TMZ and BCNU.
(A): Loss of neurosphere formation and decrease in expression of neural stem cell marker nestin in GSC11 cells after exposure to serum (10%
FBS) for the indicated time. (B): Representative of mitochondrial respiration rates in GSC11 cells cultured in stem cells medium and in serum-containing medium. Oxygen consumption was measured as an indicator of mitochondrial respiration. (C): Quantitative comparison of mitochondrial respiration rates in GSC11 cells cultured in stem cells medium and in serum-containing medium. (D): Lactate generation rates in GSC11 cells
maintained in stem cell medium or in serum-containing medium. (E): Comparison of glycolytic index in GSC11, serum-induced GSC11, GSC23, serum-induced GSC23, and glioblastoma cell line (U87). Glycolytic index was calculated according to the formula (L G)/O, in which L is the cellular lactate production, G is the glucose uptake, and O is the oxygen consumption rate. (F): Sensitivity of GSCs to TMZ, BCNU, and 3-BrOP under
hypoxic and normoxic conditions. GSC11 cells were incubated with the indicated concentrations of drugs for 72 hours in normoxia or hypoxia (2%
O2) conditions, and cell viability was measured by MTS assay. *, p < .05; **, p < .01; ***, p < .001. Abbreviations: BCNU, bis-chloroethylnitrosourea; 3-BrOP, 3-bromo-2-oxopropionate-1-propyl ester; DAPI, 4’,6-diamidino-2-phenylindole; GSC, glioblastoma stem cell; TMZ, temozolomide;
MTS, (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium); FBS, fetal bovine serum.
for 72 hours, the cells showed resistance to these conventional
chemotherapeutic drugs, especially under hypoxic conditions
(Fig. 1F). In contrast, GSCs were more sensitive to 3-BrOP
under hypoxic conditions, with the 50% inhibitory concentration of 10–15 lM compared with 20–30 lM under normoxic
conditions.
Yuan, Wang, Chen et al.
27
etry analysis after the cells were double-stained with annexin-V/
PI. Most GSC11 cells were killed after 24 hours treatment with
200 lM BCNU in combination with 15 lM 3-BrOP, whereas we
did not observe significant cell death when cells were treated
with either drug alone (Fig. 2C). The low cytotoxicity of 15 lM
3-BrOP measured by annexin-V/PI assay was likely due to the
relatively short drug incubation time (24 hours, compared to 72
hours in MTS assay). Similar synergistic drug combination effect
was also observed in GSC23 cells (data not shown). Furthermore, greater synergistic effect of BCNU and 3-BrOP was
observed in hypoxia than in normoxia, as GSCs were more dependent on glycolysis for ATP generation under hypoxia. Quantitative analysis of the drug combination index (CI) further confirmed the synergism of 3-BrOP and BCNU treatments in GSCs
under hypoxia, where the CI values were from 0.701 to 0.979
(Fig. 2D), indicating the two compounds were synergistic under
most combination conditions (CI <1, synergism; CI ¼ 1, additive; CI >1, antagonism). This finding suggests that 3-BrOP
could be an effective therapeutic agent to target GSCs especially
in hypoxic tissue environments.
Combination of 3-BrOP and BCNU Significantly
Impairs GSCs’ Sphere-Forming Ability In Vitro
and Tumorigenecity In Vivo
Since in vitro clonogenicity is considered as an indicator of the
tumor-initiating capability of cancer cells in vivo, we tested the
effect of BCNU and 3-BrOP on the ability of GSCs to form colonies of neurospheres. As shown in Figure 3A, 3B, the neurosphere-formation capacity of GSCs decreased slightly when
treated with a relatively low concentration of BCNU (50 lM) or
Figure 2. Effective killing of GSCs by combination of 3-BrOP and
BCNU. (A): Inhibition of GSC11 cells by BCNU (left panel), TMZ
(right panel), or their combination with 3-BrOP under hypoxic conditions (2% O2 for 72 hours, MTS assay). (B): Inhibition of GSC23 cells
by BCNU (left panel), TMZ (right panel), or their combination with 3BrOP under hypoxic conditions (2% O2 for 72 hours, MTS assay).
(C): Induction of cell death in GSC11 cells treated with BCNU, 3BrOP, or their combination. Cells were incubated with the indicated
concentrations of compounds for 24 hours under normoxic and hypoxic
conditions (2% O2). Cell viability was then measured using annexin-V/
propidium iodide double staining followed by flow cytometry analysis.
(D): Drug CI calculated using the Calcusyn software in GSC11 cells
treated with BCNU, 3-BrOP alone, and in combination for 24 hours
under hypoxia (2% O2). Abbreviations: BCNU (BC), bis-chloroethylnitrosourea; 3-BrOP (Br), 3-bromo-2-oxopropionate-1-propyl ester; CI,
combination index; GSC, glioblastoma stem cell; TMZ, temozolomide;
MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulfophenyl)-2H-tetrazolium; PI, propidium iodide.
Combination of 3-BrOP and BCNU Shows
Synergistic Effect in Killing GSCs, Especially
Under Hypoxic Conditions
We then tested if a combination of 3-BrOP with BCNU or TMZ
could lead to more effective killing of GSCs, based on the rationale that 3-BrOP would deplete the cellular ATP by inhibiting
glycolysis and thus impair the ability of cells to repair DNA
damage induced by BCNU or TMZ. The results of MTS assay
showed that 3-BrOP substantially potentiates the cytotoxic effect
of BCNU but not that of TMZ, especially under hypoxic conditions (Fig. 2A, 2B). We further confirmed the synergistic killing
effect of the 3-BrOP and BCNU combination, using flow cytomwww.StemCells.com
Figure 3. Impairment on GSC sphere formation in vitro and tumor
formation in vivo by combination of 3-BrOP and BCNU. (A): Representative morphology of neurosphere formation of GSCs. GSC11 cells were
incubated with the indicated concentrations of BCNU, 3-BrOP, or their
combination for 3 hours, and then cultured in drug-free medium for formation of neurospheres. (B): Quantitative data of neurosphere formation
in the presence or absence of 3-BrOP and BCNU. (C): Effect of 3-BrOP
and BCNU on tumor development in vivo. GSC11 cells were treated
with PBS (control), 20 lM 3-BrOP, 20 lM BCNU, or their combination
for 6 hours. The cells were then inoculated orthotopically into the brains
of mice (five mice in each group), and mouse survival was monitored
(without further drug treatment) as an indication of in vivo tumor formation and disease progression. Abbreviations: BCNU (BC), bis-chloroethylnitrosourea; 3-BrOP (Br), 3-bromo-2-oxopropionate-1-propyl ester;
PBS, phosphate buffered saline; GSC, glioblastoma stem cell.
28
Elimination of Glioblastoma Stem Cells
Figure 4. Preferential killing of glioblastoma cells by 3-BrOP and BCNU. The cytotoxic effects of 3-BrOP, BCNU, or their combination were
compared in (A) GSC23 cells, (B) serum (10% FBS)-induced GSC11 cells, (C) NHA, and (D) U87 glioma cells. Cells were incubated with the
indicated drugs for 24 hours under hypoxia (2% O2), and cell viability was measured by annexin-V/propidium iodide double staining followed by
flow cytometry analysis. Abbreviations: BCNU, bis-chloroethylnitrosourea; 3-BrOP, 3-bromo-2-oxopropionate-1-propyl ester; GSC, glioblastoma
stem cell; NHA, nonmalignant human astrocyte; PI, propidium iodide; FBS, fetal bovine serum.
3-BrOP (15 lM) alone. However, combination of both compounds almost completely abolished the neurosphere formation.
These results indicate that GSCs underwent irreversible cell
death and lost their self-renewal ability even after a short-term
treatment with a combination of BCNU and 3-BrOP.
We then further tested the effect of 3-BrOP and BCNU on
tumor formation in vivo by orthotopic inoculation of GSCs into
the brains of the immunodeficient mice. GSC11 cells were cultured in serum-free medium and treated with PBS (control), 20
lM of 3-BrOP, 20 lM of BCNU, or their combination for 6
hours, and then inoculated into the mouse brains. Mouse survival
was monitored as an indication of in vivo tumor formation and
disease progression. All mice in the control died within 80 days,
while all mice in the 3-BrOP- or BCNU-treated groups died with
a substantial delay, especially in the 3-BrOP-treated group (Fig.
3C), suggesting that the drug treatment partially suppresses tumor development. Importantly, only two out of the five mice in
the drug combination group died by 12 months, and the remaining three mice exhibited no signs of tumor development (Fig.
3C), suggesting that this drug combination significantly impaired
the tumorigenesis of GSCs (p ¼ .0027).
of glycolysis by 3-BrOP would preferentially deplete ATP in
GSCs and other cancer cells, but the nonmalignant cells such as
NHAs would be less sensitive to such inhibition. Indeed, as
shown in Figure 4, a combination of subtoxic concentrations of
3-BrOP (15 lM) and BCNU (200 lM) synergistically killed
most of GSC11 (Fig. 2C) and GSC23 cells (Fig. 4A). Higher concentration of 3-BrOP (20 lM) was needed to achieve similar killing effect in the serum-induced GSC11 and GSC23 cells (Fig. 4B
and Supporting Information Fig. S1). We also noted that a higher
concentration (30 lM) of 3-BrOP was needed to cause similar
killing effect in U87 cells (Fig. 4D) compared to GSC11 or
GSC23 cells (15 lM), suggesting that the GSCs are more sensitive to the combination of 3-BrOP and BCNU. Importantly, this
drug combination caused only minimum cytotoxicity in NHAs
(Fig. 4C). These observations suggest that normal human cells
may be fundamentally different from malignant cells in their
energy metabolism and sensitivity to glycolytic inhibitor, which
thus may provide a therapeutic window for selectivity. Interestingly, NHAs and U87 cells exhibited similar doubling-time but
had different drug sensitivity, while serum-induction of differentiation moderately increased the doubling-time of GSCs (Supporting Information Fig. S2) and decreased their drug sensitivity.
3-BrOP in Combination with BCNU Preferentially
Kills Tumor Cells with Relatively Low Toxic Effect
on NHAs
Severe Depletion of Cellular ATP by 3-BrOP and
BCNU in GSCs Prior to Cell Death
Since cancer cells, especially GSCs, are more dependent on glycolytic pathway to generate ATP, we hypothesized that inhibition
We then tested the effect of 3-BrOP or its combination with
BCNU on glycolysis in GSCs, using lactate production as an
Yuan, Wang, Chen et al.
29
Figure 5. Effect of 3-BrOP and BCNU or TMZ on energy metabolism in GSCs. (A): Inhibition of lactate production in GSC11 cells by the
indicated concentrations of BCNU and 3-BrOP under hypoxia (2% O2). (B): ATP depletion in GSC11 cells by the indicated concentrations of
BCNU and 3-BrOP for 3 hours under hypoxia (2% O2). (C): ATP depletion in GSC11 cells by the indicated concentrations of BCNU and 3BrOP for 6 hours under hypoxia (2% O2). (D): ATP depletion in GSC23 cells by the indicated concentrations of BCNU and 3-BrOP for 3 hours
under hypoxia. (E): ATP depletion in GSC11 cells by the indicated concentrations of TMZ and 3-BrOP for 3 hours under hypoxia (2% O2). (F):
ATP depletion in GSC23 cells by the indicated concentrations of TMZ and 3-BrOP for 3 hours under hypoxia (2% O2). (G): Viability of GSC11
cells after incubation with 3-BrOP and BCNU for 3 hours or 6 hours under hypoxic conditions (2% O2). Cell viability was measured by annexinV/propidium iodide double staining followed by flow cytometry analysis. *, p < .05; ***, p < .001. Abbreviations: BCNU (BC), bis-chloroethylnitrosourea 3-BrOP (Br), 3-bromo-2-oxopropionate-1-propyl ester; GSC, glioblastoma stem cell; TMZ, temozolomide.
indicator of glycolytic activity. As shown in Figure 5A, a 3hour incubation of GSC11 cells with 100–200 lM BCNU or
15–20 lM 3-BrOP alone caused only a slight inhibition of
lactate production. However, the combination of these two
compounds showed a significant inhibition of lactate (Fig.
5A), leading to a substantial decrease of cellular ATP by
more than 50% (Fig. 5B). A prolonged incubation with both
compounds for 6 hours in hypoxia led to significant depletion
of ATP, although incubation with 200 lM BCNU or 15 lM
www.StemCells.com
3-BrOP alone caused only a slight reduction of ATP (Fig.
5C). Similar effect of 3-BrOP and BCNU in causing lactate
production inhibition (data not shown) and ATP depletion (Fig.
5D) was also observed in GSC23 cells. In contrast, treatment
of GSCs with TMZ alone did not cause a significant change of
ATP in GSC11 or GSC23 cells, and this compound was unable
to enhance the ability of 3-BrOP to cause further depletion of
ATP in GSCs (Fig. 5E, 5F). This might explain why treatment
of GSCs with 3-BrOP plus TMZ did not show any
30
Elimination of Glioblastoma Stem Cells
Figure 6. Inhibition of GAPDH and hexokinase by BCNU and 3-BrOP. (A): Effect of BCNU and 3-BrOP on GAPDH enzyme activity in
GSC11 cells. Cells were first incubated with the indicated concentrations of BCNU or 3-BrOP for 30 minutes, and protein extracts were prepared
for analysis of GAPDH activity without further addition of the compounds in vitro. GAPDH activity was measured as described in Materials and
Methods. (B): Inhibition of purified GAPDH enzyme by BCNU in vitro. (C): Inhibition of purified GAPDH enzyme by 3-BrOP in vitro. (D): Inhibition of purified GAPDH enzyme by the combination of BCNU and 3-BrOP in vitro. (E): Effect of BCNU, 3-BrOP, or their combination on
hexokinase activity in GSC11 cells. Cells were first incubated with the indicated concentrations of compounds for 30 minutes, and protein
extracts were prepared for analysis of hexokinase activity without further addition of the compounds in vitro. (F): Inhibition of purified hexokinase II enzyme by 3-BrOP in vitro. Abbreviations: BCNU (BC), bis-chloroethylnitrosourea; 3-BrOP (Br), 3-bromo-2-oxopropionate-1-propyl
ester; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GSC, glioblastoma stem cell; HK, hexokinase.
enhancement of cytotoxicity. Flow cytometry analysis of cell
viability showed that GSCs treated with 3-BrOP plus BCNU
remained largely viable at both the 3-hour and 6-hour time
points, suggesting that the observed ATP depletion was an
early event and was not a consequence of cell death (Fig. 5G).
Effect of BCNU and 3-BrOP on the Enzyme
Activity of GAPDH and Hexokinase
The observation that BCNU but not TMZ was able to potentiate
the ability of 3-BrOP to deplete ATP led us to speculate that
BCNU might have certain unique mechanism of action different
from TMZ. Since BCNU is known to inhibit glutathione reductase by alkylating its thiolate-active site [28], we postulated that
it is possible that this compound might inhibit the glycolytic
enzyme GAPDH, which also has potential thiolate-active site. To
test this possibility, we measured GAPDH activity in GSCs
before and after treatment with BCNU or 3-BrOP alone or in
combination for 30 minutes. Interestingly, treatment with 10 lM
3-BrOP or 200 lM BCNU alone inhibited GAPDH activity by
approximately 60% (Fig. 6A), and treatment with the combination of the two compounds inhibited GAPDH activity by almost
80%, which may explain why ATP was severely depleted in the
combination setting. We further confirmed the inhibition of
Yuan, Wang, Chen et al.
purified GAPDH by BCNU and 3-BrOP at various concentrations using an in vitro enzymatic assay. The concentrations of
BCNU and 3-BrOP required to inhibit the purified GAPDH activity by 50% were approximately 200 lM and 10 lM, respectively (Fig. 6B, 6C). Incubation of purified GAPDH with the
combination of BCNU and 3-BrOP in vitro resulted in further inhibition of GAPDH activity (Fig. 6D). Interestingly, 3-BrOP (10
lM) and BCNU (100–200 lM) at the concentrations that inhibited GAPDH activity did not showed significant inhibitory effect
on hexokinase activity in GSC11 cells (Fig. 6E), suggesting that
at the low drug concentrations, the main enzyme target was
GAPDH not hexokinase. We also performed in vitro assays
using purified hexokinase II and confirmed that 3-BrOP at the
concentrations of 3–100 lM did not inhibit hexokinase activity,
while a higher concentration (300 lM) slightly inhibited hexokinase II in vitro (Fig. 6F).
3-BrOP Impairs the Ability of GSCs to Repair
BCNU-Induced DNA Damage
BCNU can cause DNA damage by formation of DNA adducts
and also cause interstrand cross-links and double-strand breaks
(DSBs) in DNA [29, 30]. DSBs are repaired in mammalian cells
by nonhomologous end-joining and homologous recombination
repair [31], in which formation of cH2AX foci is a rapid cellular
response to the presence of DSBs [32]. Because ATP is required
for H2AX phosphorylation and the subsequent complicated DNA
repair process, we speculated that depletion of ATP in GSC cells
by 3-BrOP would compromise the DNA repair. Using a comet
assay, we observed that treatment with BCNU alone or combination with 3-BrOP for 6 hours caused severe DNA damage in
GSCs, resulting in the appearance of broken DNA as the comet
tails after single-cell gel electrophoresis (Fig. 7A, 7B). In GSCs
treated with BCNU alone, the DNA damage could be repaired
3–6 hours after the removal of BCNU, as evidenced by the disappearance or reduction of the DNA tails (Fig. 7A, 7B). However, the GSCs were unable to repair the DNA damage when
treated with a combination of BCNU and 3-BrOP (Fig. 7A). In
fact, the percentage of DNA in the tails increased from 50% to
60% after a 3-hour recovery and to more than 80% after 6 hours
in GSCs treated with the drug combination. Using flow cytometry, we confirmed that the cell death in the combination treatment
was irreversible even after 18 hours of recovery in fresh medium
(Fig. 7A, lower panel).
Immediately after formation of DSBs, the MRN (MRE11,
RAD50, and NBS1) complex binds to broken DNA ends and
recruits ATM (ataxia telangiectasia, mutated), ATM- and
Rad3-related, and/or DNA-dependent protein kinase, resulting
in phosphorylation of H2AX (cH2AX) and initiation of the
DNA-repair process [33], and dephosphorylation and removal
of cH2AX in the cancer cells is a step toward turning off the
DNA damage response [34]. In our study, we observed that
when GSCs were incubated with BCNU for 4 hours alone or
in combination with 3-BrOP induced a substantial increase in
cH2AX. When we continuously treated GSCs for 6 hours, we
observed that cH2AX was present in the BCNU-treated cells
(Fig. 7C), indicating DNA damage and active repair were
ongoing. However, in cells treated with 3-BrOP or its combination with BCNU for 6 hours, cH2AX decreased substantially (Fig. 7C), suggesting a dephosphorylation of H2AX due
to a lack of ATP to support the DNA repair process.
DISCUSSION
In this study, we showed that GSCs had low mitochondrial
respiration and high glycolytic activity, which may serve as a
www.StemCells.com
31
biochemical basis for developing novel therapeutic strategies
to effectively target GSCs using a rational drug combination.
The biochemical mechanism-based drug combination of an
effective glycolytic inhibitor 3-BrOP and a conventional chemotherapeutic agent BCNU exhibited a striking synergistic
killing effect in GSCs. Almost all GSCs were killed by this
combination in 24 hours. This finding raises a potential for
use of this drug combination to eliminate GSCs and overcome
the resistance of glioblastoma to existing chemotherapeutic
agents, especially under hypoxic conditions.
Patients with GBM have poor prognosis, and there have
been only limited improvements of therapeutic outcomes during the past 2 decades [35]. GSCs seem to reside in the
niches where the cells are in hypoxic microenvironment,
which contributes significantly to cancer chemoresistance and
radioresistance [1]. Although an increase in drug dosage could
increase the effective drug concentration in the brain to kill
more cancer cells, this may also increase toxic side effect and
thus limit its applications. Furthermore, GSCs may adapt
quickly to drug treatment by upregulating the expression of
drug efflux pumps, antiapoptotic proteins, and DNA repair
proteins like O6-methylguanine-DNA methyltransferase. In
the case of BCNU, this compound has a very short half-life
and the drug become undetectable after 5 minutes of treatment [36]. BCNU penetration in the brain tissue occurs over
only a very short distance, approximately 2 mm from the ependymal surface [37]. Thus, achieving and maintaining effective high concentration of BCNU in the tumor tissue is difficult. For effective and selective killing of cancer cells,
especially CSCs, determining the fundamental differences
between cancer and normal cells is critically important.
Energy metabolism has emerged as a potential candidate target for treatment of cancers. Multiple studies have confirmed
that nutrient metabolism in cancer cells may be different from
that in normal cells. As first proposed by Warburg [38], cancer cells might rely more on glycolysis for metabolism and
survival due in part to mitochondrial dysfunction.
In this study, we observed that mitochondrial respiratory
activity was downregulated in GSCs, and glycolysis seems to
be a major energy source to maintain their survival. An effective agent that targets glycolysis can achieve the goal of eliminating CSCs. We postulated that ATP depletion induced by
glycolytic inhibition could cause a severe energy crisis, which
could lead to a severe compromise of GSC’s ability to repair
DNA damage induced by chemotherapeutic agents. However,
it seems necessary to deplete ATP to certain level in order to
effectively compromise the ability of DNA repair. In GSCs,
this severe ATP depletion could be efficiently achieved by
combination of 3-BrOP with BCNU, but not with TMZ. This
seems due to the ability of BCNU and 3-BrOP to inhibit
GAPDH activity in a synergistic fashion. Although hexokinase
has been considered a major target of 3-BrPA [19, 39-41], high
concentrations are needed to inhibit this enzyme. A recent
study using HepG2 cells suggested that this 3-BrPA might inhibit GAPDH more effectively [42]. Our study showed that 3BrOP at low concentrations preferentially inhibited GAPDH in
GSCs without significant effect on hexokinase. It should be
noted, however, that inhibition of hexokinase can still be a feasible strategy to impact cancer energy metabolism for therapeutic purpose. Interestingly, a recent study shows that suppression
of hexokinase II by siRNA causes a decrease in glycolysis and
a restoration of oxidative phosphorylation, leading to sensitization to TMZ treatment [43].
The observation that BCNU inhibited GAPDH activity is
a novel finding in study. The primary pharmacological action
of BCNU has been traditionally attributed to its alkylating activity that inhibits DNA synthesis and repair mediated by its
32
Elimination of Glioblastoma Stem Cells
Figure 7. Effect of 3-BrOP on repair of DNA damage induced by BCNU and cytotoxicity in glioblastome stem cells. (A): Comet assay of
DNA damage in glioblastoma stem cell 11 (GSC11) cells treated with BCNU, 3-BrOP, or their combination. Cells were treated with the indicated
concentrations of compounds for 6 hours, and then either immediately processed for comet assay or cultured in drug-free medium for additional
3–6 hours for potential DNA repair. The bright green dots represent the positions of cellular nuclei; the ‘‘tail’’ length and intensity and on the
right side of each nucleus represent the degree of DNA strand breaks eluted out from the cell during electrophoresis. Flow cytometry analysis
(annexin-V/propidium iodide staining) was also used to measure cell death at 24 hours (18 hours after drug removal, lower panel). (B): Quantification of DNA damage in GSC11 cells treated with or without BCNU and 3-BrOP as indicated. At least 30 cells in each sample were quantitatively analyzed for percentage of DNA tail that eluted from the cellular nuclei. (C): Western blot analysis of cH2AX and total H2AX (tH2AX)
proteins in GSC11 cells treated with the indicated compounds for 2–6 hours. *, p < .05; **, p < .01; ***, p < .001. Abbreviations: BCNU(BC),
bis-chloroethylnitrosourea; ; 3-BrOP (Br), 3-bromo-2-oxopropionate-1-propyl ester; GSC, glioblastoma stem cell; PI, propidium iodide.
active chloroethyl moieties [44]. BCNU is also known to inhibit glutathione reductase and thioredoxin reductase by alkylating its thiolate-active site [45]. Similar to glutathione reductase, GAPDH contains many thiol groups that can be
alkylated. GAPDH activity in cancer cells can be inactivated
by a number of covalent modifications, including S-nitrosylation of the protein thiols [46], disulfide formation [46], covalent binding of NADþ through a nitric oxide (NO)-dependent
thiol intermediate [47], and mono(ADP-ribosyl) ation [48].
BCNU may covalently inactivate GAPDH activity by S-nitrosylation of cysteine thiols. It is interesting to note that
although GAPDH activity could be inhibited by 50% with
200 lM BCNU in vitro using a direct enzyme assay, we
observed only a moderate decrease in cellular ATP after a 6hour treatment of GSCs with 200 lM BCNU alone. A possible explanation for the only modest ATP decrease could be
Yuan, Wang, Chen et al.
that BCNU might be less effective in inhibiting GAPDH in
intact cells, and that the moderate inhibition of glycolysis by
BCNU might have been compensated by mitochondrial oxidative phosphorylation. This is in contrast to 3-BrOP, which
seems to inhibit both glycolysis and mitochondrial respiration
[42, 49, 50]. This might explain why 3-BrOP could deplete
ATP more efficiently than BCNU when each compound was
used alone. The combination of subtoxic concentrations of 3BrOP and BCNU seems to be an effective way to inhibit
GAPDH and to achieve optimal killing of GSCs with low toxicity to normal cells. This is largely due to the unique mechanisms of action of each compound and the biological properties of GSCs, which are highly dependent on glycolysis and
thus sensitive to such metabolic intervention. The ability of 3BrOP and BCNU combination to effectively eliminate CSCs
in vitro and impair their tumor formation in vivo suggests
that this novel drug combination merits further evaluation for
their potential to be used in cancer treatment.
CONCLUSIONS
3-BrOP, a novel effective glycolytic inhibitor, can effectively
sensitize GSCs to the killing effect of BCNU, resulting in
REFERENCES
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Gilbert CA, Ross AH. Cancer stem cells: Cell culture, markers, and
targets for new therapies. J Cell Biochem 2009;108:1031–1038.
Bonnet D, Dick JE. Human acute myeloid leukemia is organized as a
hierarchy that originates from a primitive hematopoietic cell. Nat Med
1997;3:730–737.
Singh SK, Clarke ID, Terasaki M et al. Identification of a cancer stem
cell in human brain tumors. Cancer Res 2003;63:5821–5828.
Al-Hajj M, Wicha MS, Benito-Hernandez A et al. Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci USA
2003;100:3983–3988.
O’Brien CA, Pollett A, Gallinger S et al. A human colon cancer cell
capable of initiating tumour growth in immunodeficient mice. Nature
2007;445:106–110.
Li C, Heidt DG, Dalerba P et al. Identification of pancreatic cancer
stem cells. Cancer Res 2007;67:1030–1037.
Stupp R, Mason WP, van den Bent MJ et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med
2005;352:987–996.
Stupp R, Hegi ME, Mason WP et al. Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-Year analysis
of the EORTC-NCIC trial. Lancet Oncol 2009;10:459–466.
Kang MK, Kang SK. Tumorigenesis of chemotherapeutic drug-resistant cancer stem-like cells in brain glioma. Stem Cells Dev 2007;16:
837–847.
Beier D, Schulz JB, Beier CP. Chemoresistance of glioblastoma cancer stem cells—Much more complex than expected. Mol Cancer
2011;10:128.
Bao S, Wu Q, McLendon RE et al. Glioma stem cells promote radioresistance by preferential activation of the DNA damage response. Nature 2006;444:756–760.
Johannessen TC, Wang J, Skaftnesmo KO et al. Highly infiltrative
brain tumours show reduced chemosensitivity associated with a stem
cell-like phenotype. Neuropathol Appl Neurobiol 2009;35:380–393.
Liu G, Yuan X, Zeng Z et al. Analysis of gene expression and chemoresistance of CD133þ cancer stem cells in glioblastoma. Mol Cancer
2006;5:67.
Chekenya M, Krakstad C, Svendsen A et al. The progenitor cell
marker NG2/MPG promotes chemoresistance by activation of integrin-dependent PI3K/Akt signaling. Oncogene 2008;27:5182–5194.
Lei Z, Li B, Yang Z et al. Regulation of HIF-1alpha and VEGF by
miR-20b tunes tumor cells to adapt to the alteration of oxygen concentration. PLoS One 2009;4:e7629.
Heddleston JM, Li Z, McLendon RE et al. The hypoxic microenvironment maintains glioblastoma stem cells and promotes reprogramming
towards a cancer stem cell phenotype. Cell Cycle 2009;8:3274–3284.
www.StemCells.com
33
striking synergistic elimination of these cells in vitro and suppression of tumor formation in vivo. 3-BrOP is a promising
drug that preferentially targets cancer cells, especially CSCs
in hypoxic environment. Combination of 3-BrOP and BCNU
merits further evaluation in preclinical and clinical setting for
potential treatment of glioblastoma.
ACKNOWLEDGMENTS
This work was supported in part by a grant from major science
and technology project of the National Basic Research Program
(973 Program) of China (2012CB967004), and grants
CA085563, CA100428, and CA16672 from the National Institutes of Health and a grant for the CLL Global Research Foundation. Feng Wang was supported by the Rosalie B Hite
Fellowship from MD Anderson Cancer Center.
DISCLOSURE
OF POTENTIAL
CONFLICTS OF INTEREST
The authors indicate no potential conflicts of interest.
17 Li Z, Bao S, Wu Q et al. Hypoxia-inducible factors regulate
tumorigenic capacity of glioma stem cells. Cancer Cell 2009;15:
501–513.
18 Ko YH, Pedersen PL, Geschwind JF. Glucose catabolism in the rabbit
VX2 tumor model for liver cancer: Characterization and targeting hexokinase. Cancer Lett 2001;173:83–91.
19 Geschwind JF, Ko YH, Torbenson MS et al. Novel therapy for liver
cancer: Direct intraarterial injection of a potent inhibitor of ATP production. Cancer Res 2002;62:3909–3913.
20 Nelson K. 3-Bromopyruvate kills cancer cells in animals. Lancet
Oncol 2002;3:524.
21 Geschwind JF, Georgiades CS, Ko YH et al. Recently elucidated
energy catabolism pathways provide opportunities for novel treatments
in hepatocellular carcinoma. Expert Rev Anticancer Ther 2004;4:
449–457.
22 Xu RH, Pelicano H, Zhang H et al. Synergistic effect of targeting
mTOR by rapamycin and depleting ATP by inhibition of glycolysis in
lymphoma and leukemia cells. Leukemia 2005;19:2153–2158.
23 Jiang H, Gomez-Manzano C, Aoki H et al. Examination of the therapeutic potential of Delta-24-RGD in brain tumor stem cells: Role of
autophagic cell death. J Natl Cancer Inst 2007;99:1410–1414.
24 Lee J, Kotliarova S, Kotliarov Y et al. Tumor stem cells derived from
glioblastomas cultured in bFGF and EGF more closely mirror the phenotype and genotype of primary tumors than do serum-cultured cell
lines. Cancer Cell 2006;9:391–403.
25 Yu L, Baxter PA, Voicu H et al. A clinically relevant orthotopic xenograft model of ependymoma that maintains the genomic signature of
the primary tumor and preserves cancer stem cells in vivo. Neuro
Oncol 2010;12:580–594.
26 Rieske P, Golanska E, Zakrzewska M et al. Arrested neural and
advanced mesenchymal differentiation of glioblastoma cells-comparative study with neural progenitors. BMC Cancer 2009;9:54.
27 He H, Nilsson CL, Emmett MR et al. Glycomic and transcriptomic response
of GSC11 glioblastoma stem cells to STAT3 phosphorylation inhibition and
serum-induced differentiation. J Proteome Res 2010;9:2098–2108.
28 Schallreuter KU, Gleason FK, Wood JM. The mechanism of action of
the nitrosourea anti-tumor drugs on thioredoxin reductase, glutathione
reductase and ribonucleotide reductase. Biochim Biophys Acta 1990;
1054:14–20.
29 Batista LF, Roos WP, Christmann M et al. Differential sensitivity of
malignant glioma cells to methylating and chloroethylating anticancer
drugs: p53 determines the switch by regulating xpc, ddb2, and DNA
double-strand breaks. Cancer Res 2007;67:11886–11895.
30 Cui B, Johnson SP, Bullock N et al. Bifunctional DNA alkylator 1,3bis(2-chloroethyl)-1-nitrosourea activates the ATR-Chk1 pathway independently of the mismatch repair pathway. Mol Pharmacol 2009;75:
1356–1363.
31 Drablos F, Feyzi E, Aas PA et al. Alkylation damage in DNA and
RNA—Repair mechanisms and medical significance. DNA Repair
(Amst) 2004;3:1389–1407.
Elimination of Glioblastoma Stem Cells
34
32 Rogakou EP, Pilch DR, Orr AH et al. DNA double-stranded breaks
induce histone H2AX phosphorylation on serine 139. J Biol Chem
1998;273:5858–5868.
33 Ikura T, Tashiro S, Kakino A et al. DNA damage-dependent acetylation and ubiquitination of H2AX enhances chromatin dynamics. Mol
Cell Biol 2007;27:7028–7040.
34 Rai R, Peng G, Li K et al. DNA damage response: The players, the
network and the role in tumor suppression. Cancer Genomics Proteomics 2007;4:99–106.
35 Ohgaki H, Kleihues P. Population-based studies on incidence, survival
rates, and genetic alterations in astrocytic and oligodendroglial gliomas. J Neuropathol Exp Neurol 2005;64:479–489.
36 Oliverio VT. Pharmacology of the nitrosoureas: An overview. Cancer
Treat Rep 1976;60:703–707.
37 Blasberg RG, Patlak C, Fenstermacher JD. Intrathecal chemotherapy:
Brain tissue profiles after ventriculocisternal perfusion. J Pharmacol
Exp Ther 1975;195:73–83.
38 Warburg O. On the origin of cancer cells. Science 1956;123:309–314.
39 Ko YH, Smith BL, Wang Y et al. Advanced cancers: Eradication in
all cases using 3-bromopyruvate therapy to deplete ATP. Biochem
Biophys Res Commun 2004;324:269–275.
40 Xu RH, Pelicano H, Zhou Y et al. Inhibition of glycolysis in cancer
cells: A novel strategy to overcome drug resistance associated with
mitochondrial respiratory defect and hypoxia. Cancer Res 2005;65:
613–621.
41 Chen Z, Zhang H, Lu W et al. Role of mitochondria-associated hexokinase II in cancer cell death induced by 3-bromopyruvate. Biochim
Biophys Acta 2009;1787:553–560.
42 Pereira da Silva AP, El-Bacha T, Kyaw N et al. Inhibition of energyproducing pathways of HepG2 cells by 3-bromopyruvate. Biochem J
2009;417:717–726.
43 Wolf A, Agnihotri S, Micallef J et al. Hexokinase 2 is a key mediator
of aerobic glycolysis and promotes tumor growth in human glioblastoma multiforme. J Exp Med 2011;208:313–326.
44 Woolley PV, Dion RL, Kohn KW et al. Binding of 1-(2-chloroethyl)3-cyclohexyl-1-nitrosourea to L1210 cell nuclear proteins. Cancer Res
1976;36:1470–1474.
45 Witte AB, Anestal K, Jerremalm E et al. Inhibition of thioredoxin reductase but not of glutathione reductase by the major classes of alkylating and platinum-containing anticancer compounds. Free Radic Biol
Med 2005;39:696–703.
46 Molina y Vedia L, McDonald B, Reep B et al. Nitric oxide-induced
S-nitrosylation of glyceraldehyde-3-phosphate dehydrogenase inhibits
enzymatic activity and increases endogenous ADP-ribosylation. J Biol
Chem 1992;267:24929–24932.
47 McDonald LJ, Moss J. Stimulation by nitric oxide of an NAD linkage
to glyceraldehyde-3-phosphate dehydrogenase. Proc Natl Acad Sci
USA 1993;90:6238–6241.
48 Tao Y, Howlett A, Klein C. Nitric oxide regulation of glyceraldehyde-3-phosphate dehydrogenase activity in Dictyostelium discoideum
cells and lysates. Eur J Biochem 1994;224:447–454.
49 Dell’Antone P. Targets of 3-bromopyruvate, a new, energy depleting,
anticancer agent. Med Chem 2009;5:491–496.
50 Ihrlund LS, Hernlund E, Khan O et al. 3-Bromopyruvate as inhibitor
of tumour cell energy metabolism and chemopotentiator of platinum
drugs. Mol Oncol 2008;2:94–101.
See www.StemCells.com for supporting information available online.